A note on the BMO-Teichmüller space (Q892355): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
Set profile property. |
||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank |
Revision as of 01:30, 5 March 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | A note on the BMO-Teichmüller space |
scientific article |
Statements
A note on the BMO-Teichmüller space (English)
0 references
18 November 2015
0 references
The paper studies the partial topological group structure of the set of all strongly quasisymmetric homeomorphisms of the unit disk, which is denoted by \(\mathrm{SQS}(S^1)\), in the topology induced by the BMO norm. A quasisymmetric homeomorphism \(h\) is called strongly quasisymmetric if, for each \(\varepsilon>0\), there is a \(\delta>0\) such that when \(| E| \leq \delta | I| \), then \(| h(E)| \leq \varepsilon| h(I)| \), where \(I\subset S^1\) is an interval and \(E\subset I\) is a measurable subset. Strongly quasisymmetric homeomorphisms \(h\) are absolutely continuous and \(\log h'\) belongs to \(\mathrm{BMO}(S^1)\). There has been several equivalent descriptions of strongly quasisymmetric homeomorphism such as the complex dilatation of the Douady-Earle extension of a strongly quasisymmetric homeomorphism which gives a Carleson measure \(\lambda_{\mu}(z)=| \mu(z)| ^2/(1-| z| )\) on the unit disk. \(T_b=\mathrm{SQS}(S^1)/\mathrm{M\"ob}(S^1)\) is a model of the BMOA-Teichmüller space, which is a subspace of the Teichmüller space, and has been investigated by many authors because of its close relation with harmonic analysis. A partial topological group is a group with a neighborhood system at the identity which satisfies that if \(h_1\) and \(h_2\) are near to the identity, then \(h_1\circ h_2\) and \(h_1^{-1}\) are also near to the identity. It is known that the group \(\mathrm{QS}(S^1)\) of all quasisymmetric homeomorphisms of the unit disk is not a topological group, but a partial topological group in the Teichmüller metric. In the BMO-Teichmüller space, the BMO-norm introduces a new topology in \(\mathrm{SQS}(S^1)\) via \(d(h,g)=\|\log(h')-\log(g')\|_{\mathrm{BMO}}\), and this topology, called the A-Z topology in the paper, is stronger than the one induced by the Teichmüller metric. One thing which the paper shows is that \(\mathrm{SQS}(S^1)\) is a partial topological group under the A-Z topology. For a partial topological group, if the adjoint action by \(h\) is continuous at the identity, then \(h\) is said to belong to the characteristic topological subgroup. To be precise, if \(f\in \mathrm{SQS}(S^1)\) is near to the identity, then \(h\circ f\circ h^{-1}\) is also near to the identity. Gardiner and Sullivan proved that the characteristic topological subgroup of the partial topological group \(\mathrm{QS}(S^1)\) is \(\mathrm{S}(S^1)\), which consists of the mappings with vanishing ratio distortion, in the Teichmüller metric. The other thing which the paper shows is that, under the A-Z topology, the characteristic topological subgroup of \(\mathrm{SQS}(S^1)\) is \(\mathrm{SS}(S^1)\), which consists of all strongly symmetric homeomorphisms \(h\) of the unit circle, that is, of the homeomorphisms \(h\) which are absolutely continuous and such that \(\log h'\in \mathrm{VMO}(S^1)\). \(T_v=\mathrm{SS}(S^1)/\mathrm{M\"ob}(S^1)\) is a model of the VMOA-Teichmüller space. In the proof of the partial topological group structure of \(\mathrm{SQS}(S^1)\), the paper shows that the neighborhood system of the identity given by the BMO norm and the one given by the Carleson measure \(\|\lambda_{\mu}\|_C\) of \(\lambda_{\mu}=| \mu(z)| ^2/(1- | z| )\), where \(\mu\) is the complex dilatation of the quasiconformal extension of \(h\), are cofinal. Thus showing that the neighborhood system given by the Carleson measure is compatible with the group structure of \(\mathrm{SQS}(S^1)\) is equivalent to showing that the one given by the BMO norm is compatible. Concerning the Carleson measure, it is shown in the paper that \(\|\lambda_{\mu_1\circ\mu_2}\|_C\leq C(\|\lambda_{\mu_1}\|_C+\|\lambda_{\mu_2}\|_C)\) and \(\|\lambda_{\mu^{-1}}\|_C\leq C\|\lambda_{\mu}\|_C\).
0 references
quasisymmetric homeomorphism
0 references
strongly quasisymmetric homeomorphism
0 references
strongly symmetric homeomorphism
0 references
partial topological group
0 references